PROJECT SUMMARY The ability to control standing posture is an essential aspect of daily living and this ability is altered for lower-limb amputees. With the loss of active ankle joint control, amputees have decreased stability and adapt compensatory strategies of control, favoring their intact limb. A small amount of work has been done to assist postural control with a powered ankle prosthesis, though this control is ill-fitted for situations of expected perturbation due to the ability to produce only reactive responses. Normative responses involve anticipatory changes in ankle- mechanics before a perturbation occurs. Direct electromyographic (EMG) control provides promise as a means of incorporating high-level human intent via descending neural commands for the restoration of postural control mechanisms at the prosthetic ankle joint. Current work has only tested the potential for direct EMG control of a powered prosthesis with only single-muscle input or with cyclic tasks (like walking). Postural control mechanisms require coordinated control from antagonistic ankle muscles to generate patterns of reciprocal and coactivation at the ankle-joint. It is unknown then whether amputees are able to coordinate residual muscle activations and if direct EMG control with a powered ankle prosthesis can assist postural control under expected perturbation. The objective of my proposed work is to develop a deeper understanding of residual antagonistic shank muscles as a control source to help restore the ability to generate postural control mechanisms at the amputated ankle. The proposed research will facilitate my long-term goal of developing technology that will improves prosthesis functionality and amputee quality of life. Specifically, direct EMG control from residual antagonistic muscle can take advantage of human intent via descending neural commands, allowing for anticipatory modulation of prosthetic ankle-joint mechanics. Thus, I propose the following research aims: 1) To investigate residual antagonistic muscle as an input for direct EMG control; and 2) To assess the extent to which transtibial amputees use direct EMG control of a powered ankle prosthesis with antagonistic residual muscles to generate postural control mechanisms under expected perturbations. To execute my aims, I will test the ability of residual antagonist muscles in a virtual system and powered ankle prosthesis separately. I will use two virtual systems to test the ability for amputees to 1) generate varying levels of activation and coactivation and 2) control a highly dynamic system. I will quantify the ability to complete both tasks and analyze the correlation in performance. I will then use an expected perturbation testing paradigm to test the ability for amputees to use direct EMG control of a powered ankle prosthesis to enhance postural control. I will quantify the contribution of the powered prosthetic ankle to overall stability. Overall the outcomes will provide a deeper understanding of residual antagonist muscles as a control source and demonstrate the feasibility of direct EMG control for the restoration of postural control mechanisms at the prosthetic ankle. The non-invasive nature of my work also has great potential for clinical translations of direct EMG control.